Hostname: page-component-77c89778f8-m42fx Total loading time: 0 Render date: 2024-07-17T08:07:26.088Z Has data issue: false hasContentIssue false

Committee Decisions under Majority Rule: An Experimental Study*

Published online by Cambridge University Press:  01 August 2014

Morris P. Fiorina
Affiliation:
California Institute of Technology
Charles R. Plott
Affiliation:
California Institute of Technology

Abstract

This article reports the findings of a series of experiments on committee decision making under majority rule. The committee members had relatively fixed preferences, so that the process was one of making decisions rather than one of problem solving. The predictions of a variety of models drawn from Economics, Sociology, Political Science and Game Theory were compared to the experimental results. One predictive concept, the core of the noncooperative game without side payments (equivalent to the majority rule equilibrium) consistently performed best. Significantly, however, even when such an outcome did not exist, the experimental results did not display the degree of unpredictability that some theoretical work would suggest. An important subsidiary finding concerns the difference between experiments conducted under conditions of high stakes versus those conducted under conditions of much lower stakes. The findings in the two conditions differed considerably, thus calling into question the political applicability of numerous social psychological experiments in which subjects had little or no motivation.

Type
Research Article
Copyright
Copyright © American Political Science Association 1978

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

*

The financial support for this research was provided by the National Science Foundation, Economics and Political Science Programs. We gratefully acknowledge this support. For their invaluable research assistance and healthy skepticism we thank Randy Calveit, Linda Cohen, James Hong, and Darwin Niekirk. We also wish to thank John Jackson whose comments helped shape some aspects of this research.

References

1 Wilson, Woodrow, Congressional Government (New York: Meridian, 1956)Google Scholar.

2 Huitt, Ralph K., “The Congressional Committee: A Case Study,” American Political Science Review, 48 (June 1954), 340–65CrossRefGoogle Scholar; Fenno, Richard E., The Power of the Purse (Boston: Little, Brown, 1966)Google Scholar.

3 For example, in the introduction to The Power of the Purse, Fenno writes (p. xiii):

The aims of the book are threefold. In order of their likely relevance and persuasiveness for the reader they are: first, to provide an empirical description of the contemporary appropriations process in Congress; second, to demonstrate the importance of the committee-centered analysis for increasing an understanding of Congress; and third, to suggest the usefulness of certain bits of theory for students of Congress and its committees.

4 Plott, Charles R., “Axiomatic Social Choice Theory: An Overview and Interpretation,” American Journal of Political Science, 20 (August 1976), 511–96CrossRefGoogle Scholar; Riker, William and Ordeshook, Peter, An Introduction to Positive Political Theory (Englewood Cliffs, N.J.: Prentice-Hall, 1973), passimGoogle Scholar; Fishburn, Peter, The Theory of Social Choice (Princeton, N.J.: Princeton University Press, 1973)Google Scholar; Ferejohn, John and Fiorina, Morris, “Purposive Models of Legislative Behavior,” American Economic Review Proceedings and Papers, 65 (05 1975), 407–14Google Scholar.

5 An occasional working adult (e.g., gardener, secretary) appeared among our subjects. Our subjects included members of racial minorities and women. We are aware that some social psychologists prefer subject populations to be racially or sexually homogeneous. But the class of models in which we are interested gives us no reason to differentiate among blacks, whites, chicanos, men and women.

6 This restriction serves two functions. First, it enhances control over the experiment. Outside deals which induce preference changes in subjects but not in the models are made difficult to negotiate. Furthermore, social stigmas regarding monetary endowments which also can induce uncontrolled preferences are minimized when monetary amounts are not public. Secondly, and perhaps more importantly, a cardinal measure of returns is not a general property of a broad range of committee processes. As in our experiments political decision makers can leam the order in which their fellow decision makers regard alternative proposals. They can attempt to communicate intensity through anger, other displays of emotion, and effort. But objectively given cardinal measures of returns are generally absent.

7 For all (x 1, X 2) and (x1, x2) we know that (x 1, x 2)Ri(x1, x2) if and only if Ui(x 1, x 2) ≥ Ui ≥ (x1, x2). So the binary relation Ri is the preference relation.

8 It is interesting to compare this stopping rule with that used by Halfpenny, Peter and Taylor, Michael, “An Experimental Study of Individual and Collective Decision Making,” British Journal of Political Science, 3 (1973), 425–44CrossRefGoogle Scholar. In their experiments Halfpenny and Taylor operationally defined an equilibrium as a 5-time winner, i.e., the experiment ended when a motion on the floor had defeated five alternative motions. In our Series 1 experiments 24 proposals in 17 experiments defeated 5 or more alternatives but did not go on to become the group decision. Thus, 17 of 40 experimental outcomes would have been different had we taken action to stop the experiments rather than allowing them to take their course (and assuming such a rule left individual behavior unchanged). This example graphically illustrates the potential impact of procedural variations on experimental outcomes.

9 Aumann, R. J., “A Survey of Cooperative Games Without Side Payments,” in Essays in Mathematical Economics in Honor of Oskar Morgenstern, ed. Shubik, M. (Princeton, N.J.: Princeton University Press, 1967)Google Scholar.

10 Aumann.

11 Black, Duncan, The Theory of Committee and Elections (Cambridge: Cambridge University Press, 1958)Google Scholar; Black, Duncan and Newing, R. A., Committee Decisions with Complementary Valuations (London: William Hodge, 1951)Google Scholar; Plott, Charles R., “A Notion of Equilibrium and Its Possibility under Majority Rule,” American Economic Review, 57 (September 1967), 797806Google Scholar; Sloss, Judith, “Stable Outcomes in Majority Rule Voting Games,” Public Choice, 15 (Summer 1973), 1948CrossRefGoogle Scholar.

12 To draw a theoretical distinction between the formally equivalent concepts of core and voting equilibrium might strike some of our colleagues as unnecessary hair-splitting. To these skeptics we address one further point. In the typical case in which no core/equilibrium exists, where does one turn? If one has arrived at the core through the game-theoretic literature, one naturally turns to some other game-theoretic concept such as the N-M solution. But if one has arrived at the equilibrium through the voting-theoretic literature, one turns to some other voting-theoretic model such as the min-max set or an agenda theory. When it exists, the core/equilibrium is the intersection of several branches of theory. But that is not to deny that these branches are distinct.

13 Simpson, Paul B., “On Defining Areas of Voter Choice,” The Quarterly Journal of Economics, 83 (08 1969), 478–90CrossRefGoogle Scholar.

14 Rae, Douglas and Taylor, Michael, “Decision Rules and Policy Outcomes,” British Journal of Political Science, 1 (1971), 7190CrossRefGoogle Scholar; also, Halfpenny and Taylor, “An Experimental Study.”

15 Charles R. Plott and Michael E. Levine, “On Using the Agenda to Influence Group Decision: Theory, Experiments and Application,” American Economic Review, forthcoming.

16 Plott and Levine, “On Using the Agenda.”

17 McKelvey, Richard, “Intransitivities in Multi-dimensional Voting Models and Implications for Agenda Control,” Journal of Economic Theory, 12 (1976), 472–82CrossRefGoogle Scholar.

18 C. R. Plott and Michael E. Levine, “On Using the Agenda.”

19 Riker, William, A Theory of Political Coalitions (New Haven: Yale University Press, 1962)Google Scholar.

20 Gamson, William, “A Theory of Coalition Formation,” American Sociological Review, 26 (1961), 373–82CrossRefGoogle Scholar; Komorita, S. S. and Chertkoff, Jerome M., “A Bargaining Theory of Coalition Formation,” Psychological Review, 80 (05 1973), 149–62CrossRefGoogle Scholar.

21 Obviously, our procedures (no communication of monetary amounts) work against any theory which specifies that the committee or a subgroup of it tries to achieve a group maximum (models 12 and 13). Of course, the political world also works against any such theory in that cardinal measures of preferences which group maximization theories presuppose are typically not available (footnote 3). And even if objectively known monetary payoffs are available, there still is no theoretical justification for asserting the equivalence of a maximum of group monetary payoff and group utility payoff. But for reasons discussed under model 13 we can not dismiss out-of-hand models which contain group or subgroup maximization processes.

22 Lieberman, Bernhardt, “Combining Individual Preferences into a Social Choice,” in Social Choice, ed. Lieberman, Bernhardt (New York: Gordon and Breach, 1971), pp. 525Google Scholar.

23 Hanson, R. A. and Rice, P. M., “Policy Outcomes in Decision Making Groups,” paper delivered at the annual meeting of the American Political Science Association, 1972Google Scholar.

24 Schelling, Thomas, The Strategy of Conflict (Oxford: Oxford University Press, 1960)Google Scholar.

25 Interestingly enough, this seeming altruism has an ironic twist. Subjects often made the incorrect assumption that their payoffs were identical. But the asymmetry in payoff structures (Figure 2) leads to the result that if Players 1,2,3 heeded the pleas of Players 4 and/or 5 to move farther north and/or east than the equilibrium, then the fringe players got significantly more in absolute terms than the more centrally located players, although the outcome remained relatively closer to the latter.

26 Fishburn, Peter, “Paradoxes of Voting,” American Political Science Review, 68 (06 1974), 537–46CrossRefGoogle Scholar.

27 McKelvey, “Intransitivities in Spatial Voting Games.”

28 Tullock, Gordon, “The General Irrelevance of the General Impossibility Theorem,” Quarterly Journal of Economics, 81 (05 1967), 256–70CrossRefGoogle Scholar.

29 There are various game-theoretic solution concepts (N-M Solution, Bargaining Set, Competitive Solution) not as yet computed which also remain live candidates.

30 cf. Rapoport, Anatol and Chammah, Albert, Prisoner's Dilemma (Ann Arbor: University of Michigan Press, 1965)CrossRefGoogle ScholarPubMed; Peter Halfpenny and Michael Taylor, “An Experimental Study of Individual and Collective Decision Making.”

31 Actually, each of the three arguments discussed in this section is a member of a more general class which holds that the laboratory environment is artificial (the word is usually pronounced loudly and repeatedly). The specifics of the general criticism often are not spelled out–we have tried to do so in this section–but at base the criticism appears to stem from acceptance of a gestalt psychology view. That is, the subtleties of the empirical context are too many and/or too complicated to be enumerated. Moreover, these variables may interact in such a way that they are effective only when moved together–precisely the kind of variation not allowed in the controlled laboratory environment. This view must be taken seriously, and ultimately there is only one convincing rejoinder, the same that an experimentalist in the physical sciences must give to a critic who claims that the laws which operate in test tubes or other laboratory environments differ from those which operate in the “real world.” One can only point to laboratory-generated information which has been helpful in field or engineering endeavors. Granted, experimental social science has not yet progressed to such a level, but we are optimistic that it eventually will.

* page 594 This section omitted in no-communication condition.

Submit a response

Comments

No Comments have been published for this article.